-> Validated score still wrong? Let us know here

[knopflerbruce]

http://www.hwbot.org/result.do?resultId=802259

 

AMD is back on the TOP!!1!

 

No, seriously... please move this one to 32m:p

[knopflerbruce]

http://www.hwbot.org/result.do?resultId=785877 - seems to be a wrong pifast version. Memory amount is not what it usually is, and the score itself can't be achieved with this chip running at 2640 MHz

[r1ch]

http://www.hwbot.org/result.do?resultId=589614 - FM link broken?

http://www.hwbot.org/result.do?resultId=658677 - no CPU-Z?

http://www.hwbot.org/result.do?resultId=676877 - resolution blocked?

http://www.hwbot.org/result.do?resultId=658673 - no CPU-Z?

http://www.hwbot.org/result.do?resultId=638110 - no CPU-Z/GPU-Z/Sub scores not shown

http://www.hwbot.org/result.do?resultId=657566 - gpu-z blocked, no cpu-z, no sub scores

 

Is there a reason that demiurg has allowed some of these results after I reported them the first time without them satisfying the submission criteria? I don't understand.

[r1ch]

I don't want to have a go at hwbot, I think you guys do a fantastic job here and I am very grateful.

 

From the rules:

Any online 3DMark06 submission must comply to the following rules:

use default 3Dmark settings

have a valid screenshot (see example below): clearly show 3Dmark score, 3Dmark subtest scores, 3Dmark settings, processor in CPU-Z, videocard in GPU-Z,

 

This is very clear, but what is the point if the real rule is that the score is "not suspicious"?

 

If a screenshot without a CPU-Z window is ok, why is it required? And worse still, if these results are accepted, what is to stop someone running a bench at default settings, photoshopping the points to a higher score and then typing in the requred cpu/gpu mhz and approriate cooling method to make the score 'not suspicious'?

 

I am not accusing anyone of cheating, I am asking why there are rules that don't need to be followed.

[komadyret]

Rules have changed through the years...

 

But I do agree with you r1ch. There should be a posibilty to have SOME verification up even after a cretain time, and a screenie/verification that is not in compliance with the rules AT TIME OF POSTING should be taken down, regardless of age and if the score is "suspicious" or not.

 

Also, I agree with the mods. If a score isn't questioned after 6 months, I should be able to take down the orb-link (verification) and free up some space for fresh runs with new hardware for the ORB. A screenshot uploaded tho, are down to the hwbot maintainers discretion to remove.

[knopflerbruce]

Rules have changed through the years...

 

But I do agree with you r1ch. There should be a posibilty to have SOME verification up even after a cretain time, and a screenie/verification that is not in compliance with the rules AT TIME OF POSTING should be taken down, regardless of age and if the score is "suspicious" or not.

 

Also, I agree with the mods. If a score isn't questioned after 6 months, I should be able to take down the orb-link (verification) and free up some space for fresh runs with new hardware for the ORB. A screenshot uploaded tho, are down to the hwbot maintainers discretion to remove.

 

If the GPU you used for that 6 months old submission is not so common, it's a good chance no-one will spot the faults - even after six months. If more than 20 people have benched the card, I'd say it's a fair chance some of those 20 would've checked the other results to see if they were OK, but with rare stuff this doesnt happen much. If you bench only the popular cards, maybe that's a good way of freeing up ORB space, but with those rare ones... I don't like it. But if that's the general rule of thumb here, I really hope people will upload their screenshots so the bot won't be full of unverifiable submissions.

[r1ch]

I understand what you're saying jmke, thanks for explaining the mods perspective on this.

 

However.

 

I understand the results aren't recent, but does this matter? I agree the scores aren't 'out of line' but the required parts of the screenshot, even for 1 year old results, aren't there. That means that there can be results after 6 months that come up as questionable.

 

The one that i'd suggest creates the most debate (the rest are me being a bit picky) is this one:

http://www.hwbot.org/result.do?resultId=638110

 

There's no proof at all there. I'm not saying it's a cheat, but there's just no justification of anything. It's not about this one result, it's whether a result like this should be allowed to stand.

 

Sorry to drag this on, this is the last I'll post on it so it doesn't carry on. Just wanting to create the point of discussion so it's clear for everyone what the rules are.

 

Thanks

[Narmer]

Could you check this result:

http://www.hwbot.org/result.do?resultId=797272

 

My vs his:

T7100 1.8GHz vs T7100 1.8GHz

GeForce 8600M GT 685/475 vs GeForce 8600M GT 670/470

13600 vs 17675

huh?

 

My run:

http://www.hwbot.org/result.do?resultId=700940

Edited December 28, 2008 by Narmer

[fakesnake]

hi just had an old result blocked, think I have fixed it now proof was not showing.

validation link now works, can it be checked and put back please.

thank you.

 

http://www.hwbot.org/result.do?resultId=768834

[Narmer]

you got a comparison link?

http://service.futuremark.com/resultComparison.action?compareResultId=5706108&compareResultType=10

this is his, stock settings; may be a bugged run, don't know enough of 3D03 to judge

 

My run:

http://www.hwbot.org/result.do?resultId=700940

[fakesnake]

fix0red; scores with links to the old x86 site shouldn't be blocked in the first place

 

thank you, every point helps.

[r1ch]

what can we make from this screenshot?

- total score

- 3DMark03 free version = standard settings only

- VGA used

- CPU speed

 

quite a bit of info there; and it not really out of line or impossible score either; and it scores not even 1HWboint?

 

I know I said I wouldn't post again, but I think you missed my point so this is my final reply.

 

Look at the things you definitely don't know, which could contribute to a huge point difference, probably +/- 50%.

 

i.e

AMD/Intel, model, cache.

RAM and timings.

GFX core/mem clocks, driver used, SLI?

Sub-test scores.

 

My main point it, it's not about the score, whether it's out of line, or how many points it got. It's about that screenshot being 'enough' for a score that doesn't appear to be 'out of line', which I don't think it is.

 

Thanks for your replies

[komadyret]

I'm with you on that r1ch. I don't think a score should be left standing if it obviousy didn't meet the requirements for a valid submission at the time of posting. Not even if the score has been left unnoticed for months/years.

 

The score in question don't show any proof that it was run at default resolution? It may be defaulting to a different resolution beeing a mobile gpu and all (hardware limitation), despite beeing a unregistered 3dmark03 copy?

It is not even showing the full screenie, to hint at the resolution-capabilities of the lcdpanel.

[K404]

http://www.hwbot.org/compare.do?resultId=748081

 

report option is grayed out but hes using GPU-based PhysX

 

(can be seen on his FM link...he doesnt have a screeny of the score present)

[r1ch]

http://www.hwbot.org/result.do?resultId=689732'>http://www.hwbot.org/result.do?resultId=689732

http://www.hwbot.org/result.do?resultId=673816

http://www.hwbot.org/result.do?resultId=693269

http://www.hwbot.org/result.do?resultId=677496

http://www.hwbot.org/result.do?resultId=689732

 

Blocked before AM3 rules updated.

 

Sorry I don't know if they'll score points or not now, but should be unblocked. Thanks.

Edited January 3, 2009 by r1ch

[CaNNon]

13 May 2008 10:08 insufficient verification (crew) jmke insufficient verification for rankings (Since 17/10/2007 use of GPU-Z is mandatory for 3D benchmarks screenshots http://hwbot.org/hwbot.post.do?postId=717)

 

he does show the rule date in that one, is it not correct?

[r1ch]

he does show the rule date in that one, is it not correct?

 

He does, but this one overrides it.

 

http://hwbot.org/hwbot.post.do?postId=872

 

And the first comment by jmke:

 

"if you have an older score (before 15/5) which was blocked for this reason; please don't hesitate to contact the mods in this thead: http://www.hwbot.org/forum/showthread.php?t=1787 posting the URL of the score that was blocked; thanks!"

 

[CaNNon]

Looks like thats addition of cpu-z screen shot too, a little confusing especially if they didn't add both at the same time.

[Narmer]

Hey, becouse now hwbot engine is strange, I dont have contact with user RomanLV from Overсlockers.сom.UA Team. Anyone who has contact with him, please show him this post.

 

I want more proof to results of Radeon 9550, more specification and descrition, model of card, photos etc. Thanks in advance.

 

http://hwbot.org/quickSearch.do?hardwareId=GPU_343

[Sweet]

Hey, becouse now hwbot engine is strange, I dont have contact with user RomanLV from Overсlockers.сom.UA Team. Anyone who has contact with him, please show him this post.

 

I want more proof to results of Radeon 9550, more specification and descrition, model of card, photos etc. Thanks in advance.

 

http://hwbot.org/quickSearch.do?hardwareId=GPU_343

 

Yes NARMER , I Think this submit score was wrong

 

RV 350 is for Radeon 9600, and RV 360 is for 9550 Core., I understand that, unless something is not known, or something new:rolleyes:

 

All score of RomanLV is RV 350, not Rv 360.

 

http://www.hwbot.org/result.do?resultId=807783

 

http://www.hwbot.org/result.do?resultId=807776

 

http://www.hwbot.org/result.do?resultId=807792

 

http://www.hwbot.org/result.do?resultId=807797

 

http://www.hwbot.org/result.do?resultId=807799

 

In all the screen's reads core 350

 

Regards

 

Sw

Edited January 4, 2009 by Sweet

[Narmer]

Sweet

Thx for reply and understanding, I dont think RomanLV's scores are cheated, but this results are very good for his platform (Athlon 64 etc), so I want to see card model, photos etc.

 

So this results should be deleted?

[71proste]

These results should be moved to correct category

[Sweet]

So this results should be deleted?

 

No, I think not delete, this result are valid, but moved to correct category

 

These results should be moved to correct category

 

Is what I say in my bad English, sorry, but you understand me...

 

Well , my spanish is better

 

bye guys

 

Sw

[Massman]

Now we have

[Narmer]

No, I think not delete, this result are valid, but moved to correct

So what with this?